The default is 0.1 M glycine-HCl at pH 3.0, applied over 3–5 column volumes and collected into 1 M Tris-HCl at pH 8.5–9.0 — about 30–50 µL per mL of fraction for a glycine pool, aiming for a pooled pH of 6.5–7.5. It works for most antibodies, and it works the same way on a packed bed and on a MonoCore™ Protein A capsule, where the 3–5 column volumes pass in seconds at about 8 column volumes per minute. If your product aggregates, the first thing to change is not the buffer but the time it spends at low pH; the second is the additive, and only then the buffer species — acetate or citrate at a slightly higher pH is usually the better trade.
The four buffers in practice
| Buffer | Typical composition | Working pH | What it is good at | Watch out for |
|---|---|---|---|---|
| Glycine-HCl | 0.1 M glycine, pH adjusted with HCl | 2.7 – 3.0 | Cheapest and most predictable; buffers well up to about pH 3.2 | Above pH 3.2 its capacity falls away — at pH 3.5 it holds roughly 40 % of what it holds at 3.0, so if you want a milder elution the species has to change. It is harsh because of the pH it is used at, not because of the molecule |
| Acetic acid / acetate | 1 M acetic acid, or 50–100 mM sodium acetate, pH 3.5–4.0 | 3.0 – 4.0 | Well poised at the upper end of the elution window and for a low-pH hold; volatile in the acetic-acid form | Sodium acetate is not volatile — the sodium stays. At 0.1 M and pH 3.0, acetic acid holds about a sixth of the capacity of 0.1 M glycine at the same pH; its pKa is 4.76, so use the 1 M form or a titrated acetate buffer |
| Citrate | 50–100 mM sodium citrate | 3.0 – 3.5 | Flat buffer capacity across the whole window; at 100 mM it out-buffers glycine and acetate, at 50 mM it is on par with 0.1 M glycine at pH 3.0 | Chelates metal ions, and it needs considerably more base to neutralise than a glycine pool of the same molarity. Reports on aggregation disagree — see below |
| Arginine (additive, not a buffer) | 0.5–2 M arginine in acetate or glycine, which sets the pH | 4.0 – 4.5 | Releases antibody at a pH where a plain buffer no longer elutes — the guanidinium group weakens the Fc–ligand interface rather than shifting the pH | Very high ionic strength after titration to pH 4–4.5: the pool usually needs dilution or buffer exchange before an ion exchange step. Viscosity and cost on top. It does not suppress aggregation in every case |
The literature does not hand you a ranking. In a 2025 study on an IgG4 biosimilar, 100 mM sodium citrate at pH 3.5 came out favourably — and the authors state explicitly that the relative merit of glycine and citrate varies with the molecule and the pH. A 2025 hold-stability study put the order the other way round, with glycine the most stable and citrate the least. The larger effect in the first study came from neither buffer: an additive did it, with mannitol cutting high-molecular-weight species from about 15 % to under 2 %, and polysorbate 80 and magnesium chloride also improving the monomer fraction. Screen the additive before you argue about the acid.
Hold time beats buffer choice
Aggregation at low pH is a kinetic process. The same antibody in the same buffer will aggregate far more in twenty minutes than in two, which is why the way you collect matters more than which acid you chose. The levers, in the order they actually matter: neutralise in the collection tube rather than after pooling; elute at the highest pH that still desorbs the product, which is a question of ligand as much as of buffer; choose the buffer and, more importantly, the stabilising additive; and only then the residence time of the step itself, which on a MonoCore™ Protein A capsule passes in seconds at about 8 column volumes per minute.
- Neutralise in the collection tube, not after pooling. Pre-load each tube with 1 M Tris-HCl at pH 8.5–9.0, sized for the acid you are putting in: about 30–50 µL per mL of expected fraction volume for a 0.1 M glycine pool, 120–250 µL for citrate or acetate.
- Collect in smaller fractions during the peak, so no fraction sits at low pH while the rest of the peak is still eluting.
- Check the pH of the pooled, neutralised product — the target is a return to near-neutral, and glycine-rich pools can land lower than you expect.
- If a low-pH hold is required for viral inactivation, that is a separate, deliberately timed step. Do not let it happen by accident in the collection tube.
Verify by measuring aggregate at the elution condition, not only after neutralisation. Size-exclusion analysis of the neutralised pool tells you what survived; it does not tell you what the acid did.
Getting the neutralisation ratio right
The rule of thumb is buffer-specific, and the common “60–200 µL of 1 M Tris per mL” only fits one case. What you are aiming for is a pooled pH of 6.5–7.5, and the base you need follows from the acid you put in: about 30–50 µL of 1 M Tris-HCl per mL of fraction for 0.1 M glycine at pH 2.7–3.0; roughly 120–250 µL for 50–100 mM citrate or acetate; and for 1 M acetic acid, 1 M Tris-HCl is the wrong reagent altogether — that pool needs Tris base or a 2 M stock. Take the ratio as a starting point and verify the result, not the other way round.
Determine it once, properly: run a cycle with buffer only, collect into the pre-loaded tubes, and measure the pH of each fraction. Then repeat the check on your first real eluate, because the product buffers too. Measure at the temperature you work at — the pKa of Tris shifts by about 0.028 units per degree, so a cold pool reads higher than the same pool at room temperature. If any fraction lands below pH 5, increase the base volume before you put material you care about through it.
The exception: low-pH virus inactivation
Everything above assumes you want the antibody out of the acid as quickly as possible. A bioprocess has one deliberate exception: the Protein A pool is also where enveloped viruses are inactivated. The generic conditions are pH 3.6 or below — in practice pH 3.4–3.6 — held for 30 to 60 minutes at 14–25 °C, with pH, time and temperature validated per molecule. Standard practice is to titrate the pool down into that window with acetic or citric acid, which is one more argument for eluting with acetate or citrate close to pH 3.4–3.6 rather than with glycine at 3.0: a glycine pool is poorly poised at 3.6, and raising a pH 3.0 pool with base risks local overshoot in a step whose claim rests on the whole pool having stayed below the limit. The step itself is described in the Protein A column protocol.
When the acid itself is the problem
Some formats do not tolerate the step at all. Bispecifics, ADCs and Fc-fusions are the usual candidates, and no buffer choice fully solves it — the mechanism and the options are set out in why low-pH elution fails on sensitive antibody modalities. For these cases we work on a different route: Solaris® Protein A binds under red light (630 nm) and elutes under blue light (480 nm) at neutral pH. It is in Beta, with access by application — not a released product, and not an answer for a process you need to run next week.
What if the answer is to avoid the acid altogether?
Everything above assumes the elution is acidic and asks which acid to use. For a molecule that does not tolerate the step at all, the question changes: engineered ligands release around pH 4.3–5.0, amino-acid buffer systems carry the pool back towards neutral, and a few routes — calcium-dependent ligands, detergent elution, light-controlled release — remove the acid entirely. All nine are compared, with what each one costs, in how to elute from Protein A without acid.
Frequently asked questions
What pH does Protein A elution need?
Release generally happens between pH 2.5 and 3.5, depending on the ligand and the antibody subclass. Start at pH 3.0 and move up if the product allows it: every tenth of a pH unit you can give back reduces the aggregation risk. Confirm the actual pH of the prepared buffer with a meter — nominal values drift with temperature and dilution.
Glycine or acetate for Protein A elution?
Glycine if you want the cheapest, most predictable release and your product tolerates pH 2.7–3.0; it stops buffering usefully above about pH 3.2. Acetate or citrate if you want to work at pH 3.4–3.6, which is both gentler on many products and closer to the virus inactivation window. Which of the two is kinder to your molecule is molecule-specific — the published comparisons disagree.
How much Tris do I add to neutralise?
Enough 1 M Tris-HCl at pH 8.5–9.0 to land the pool between pH 6.5 and 7.5 — about 30–50 µL per mL of fraction for 0.1 M glycine, 120–250 µL for 50–100 mM citrate or acetate, and a different reagent entirely for 1 M acetic acid, which needs Tris base or a 2 M stock. Pre-load the tubes, then verify the result once with buffer only and once on real eluate, because the product buffers too.
Can I elute Protein A at neutral pH?
Not with a classical Protein A ligand on its own: release depends on the pH-driven conformational change. There are ways around it — arginine with a co-solvent such as propylene glycol has been reported to release antibody at neutral pH, and ligands engineered for elution at pH 4.0–4.5 or for salt-mediated elution shift the working point without additives. The more fundamental route is a ligand that releases on a different trigger altogether, which is what light-controlled affinity is about.
Does the low-pH hold for virus inactivation conflict with aggregation?
It is the same acid exposure, used on purpose. Manage it by validating the mildest pH that still inactivates, by controlling temperature and time rather than extending them "for safety", and by measuring aggregate at the end of the hold. If a format cannot survive a validated hold, the inactivation strategy — not the elution buffer — is what has to change.
Does the elution buffer affect the next step?
Yes. Citrate chelates metals and is harder to remove; arginine adds viscosity and has to be cleared; acetate is volatile and forgiving. If the next step is ion exchange, conductivity after neutralisation decides whether you need a dilution or a buffer exchange.
Why does my antibody precipitate when eluted into Tris?
Because the pool passes through the product's isoelectric range, or sits there while more acid arrives. Pre-load the neutralisation buffer so each fraction is neutralised as it lands, collect smaller fractions, and check that the pooled pH ends between 6.5 and 7.5. For a molecule that keeps doing it, move the elution to a milder pH with citrate or add arginine — and verify aggregate by size exclusion at the elution condition, not only after neutralisation.
Still an open question?
Tell us the molecule, the feed volume and the system you run it on, and we will say whether a MonoCore™ capsule is the right starting point — or whether it is not.